Artificial intelligence self-learning-based static electrocardiography analysis method and apparatus

1. An artificial intelligence self-learning-based resting electrocardiogram analysis method, comprising:

receiving resting electrocardiogram data output by a resting electrocardiogram monitoring device, and converting a data format of the resting ECG data into a preset standard data format by resampling, and performing a first filtering processing on converted resting ECG data in the preset standard data format;

performing heart beat detection processing on resting ECG data processed by the first filtering processing to identify multiple pieces of heart beat data comprised in the resting ECG data, each of which is corresponding to a heart beat cycle, comprising amplitude data and starting-ending time data of corresponding P wave, QRS complex and T wave; and lead parameters of the heart beat data are determined;

combining the heart beat data to generate heart beat time sequence data based on the lead parameters and time rules; and generating heart beat analysis data according to the heart beat time sequence data;

performing feature extraction and analysis of an amplitude and time characterization data on the heart beat analysis data according to a trained heart beat classification model, to obtain primary classification information of the heart beat analysis data;

inputting the heart beat analysis data of particular heart beats in results of the primary classification information into a trained ST segment and T wave change model for identification, and determining ST segment and T wave evaluation information;

performing secondary classification processing on the heart beat analysis data according to resting ECG basic rule reference data, detailed feature information of the P wave and the T wave and the ST segment and T wave evaluation information under the primary classification information to finally obtain heart beat classification information;

according to the heart beat time sequence data and the heart beat classification information, performing heart beat waveform feature detection on the heart beat analysis data to determine feature heart beats; determining detailed feature information of the QRS complex in the feature heart beats; determining the detailed feature information of the P wave and the T wave in the feature heart beats; and calculating average waveform data and measurement values of the feature heart beats; and

for data of the heart beat classification information and the average waveform data of the feature heart beats, generating and outputting ECG event data, data of the measurement values and the average waveform data of the feature heart beats according to the resting ECG basic rule reference data and a classification standard of ECG Minnesota coding.

2. The resting electrocardiogram analysis method according to claim 1 , wherein the generating heart beat analysis data according to the heart beat time sequence data comprises:

cutting the heart beat data of each lead in the heart beat time sequence data according to a preset threshold to generate the heart beat analysis data of each lead.

3. The resting electrocardiogram analysis method according to claim 2 , wherein the determining the detailed feature information of the P wave and the T wave in the feature heart beats comprises:

performing QRS complex signal elimination processing on the heart beat analysis data, and performing second filtering on the heart beat analysis data after the QRS complex signal elimination processing, performing data separation on the heart beat analysis data by an independent component analysis algorithm to obtain sequence data of each independent component, according to distribution characteristics of peak values of the sequence data of the independent component and a position of the QRS complex, selecting an independent component with the highest probability as corresponding P wave and T wave components, and determining direction and morphology features of the P wave and the T wave.

4. The resting electrocardiogram analysis method according to claim 3 , wherein the performing feature extraction and analysis of an amplitude and time characterization data on the heart beat analysis data according to a trained heart beat classification model, to obtain primary classification information of the heart beat analysis data comprises:

according to a trained lead synchronous correlation classification model, performing the feature extraction and analysis of a synchronous amplitude and the time characterization data on the heart beat analysis data of each lead with a first data amount, to obtain the primary classification information of the heart beat analysis data.

5. The resting electrocardiogram analysis method according to claim 4 , wherein the inputting the heart beat analysis data of particular heart beats in results of the primary classification information into a trained ST segment and T wave change model for identification, and determining ST segment and T wave evaluation information comprises:

inputting data of the particular heart beats in the primary classification into the trained ST segment and T wave change model according to leads in turn, performing the feature extraction and analysis of the amplitude and the time characterization data on the data of the particular heart beats of each lead to obtain ST segment and T wave change information of each lead, and determining the ST segment and T wave evaluation information, which is lead position information that indicates the ST segment and T wave corresponding to heart beat segment data occur change.

6. The resting electrocardiogram analysis method according to claim 5 , wherein the method further comprises:

receiving modification information of the heart beat classification information; and

taking modified data as training sample data for model training in the artificial intelligence self-learning-based electrocardiogram automatic analysis method.

7. A computer program product comprising instructions, wherein when the computer program product runs on a computer, the computer executes the method of claim 6 .

8. A computer readable storage medium, comprising instructions, wherein when the instructions run on a computer, the computer executes the method of claim 6 .

9. The resting electrocardiogram analysis method according to claim 1 , wherein the determining the detailed feature information of the P wave and the T wave in the feature heart beats comprises:

performing QRS complex signal elimination processing on the heart beat analysis data, and performing second filtering on the heart beat analysis data after the QRS complex signal elimination processing, performing data separation on the heart beat analysis data by an independent component analysis algorithm to obtain sequence data of each independent component, according to distribution characteristics of peak values of the sequence data of the independent component and a position of the QRS complex, selecting an independent component with the highest probability as corresponding P wave and T wave components, and determining direction and morphology features of the P wave and the T wave.

10. The resting electrocardiogram analysis method according to claim 1 , wherein the performing feature extraction and analysis of an amplitude and time characterization data on the heart beat analysis data according to a trained heart beat classification model, to obtain primary classification information of the heart beat analysis data comprises:

according to a trained lead synchronous correlation classification model, performing the feature extraction and analysis of a synchronous amplitude and the time characterization data on the heart beat analysis data of each lead with a first data amount, to obtain the primary classification information of the heart beat analysis data.

11. The resting electrocardiogram analysis method according to claim 1 , wherein the inputting the heart beat analysis data of particular heart beats in results of the primary classification information into a trained ST segment and T wave change model for identification, and determining ST segment and T wave evaluation information comprises:

inputting data of the particular heart beats in the primary classification into the trained ST segment and T wave change model according to leads in turn, performing the feature extraction and analysis of the amplitude and the time characterization data on the data of the particular heart beats of each lead to obtain ST segment and T wave change information of each lead, and determining the ST segment and T wave evaluation information, which is lead position information that indicates the ST segment and T wave corresponding to heart beat segment data occur change.

12. The resting electrocardiogram analysis method according to claim 1 , wherein the method further comprises:

receiving modification information of the heart beat classification information; and

taking modified data as training sample data for model training in the artificial intelligence self-learning-based electrocardiogram automatic analysis method.

13. A computer program product comprising instructions, wherein when the computer program product runs on a computer, the computer executes the method of claim 1 .

14. A computer readable storage medium, comprising instructions, wherein when the instructions run on a computer, the computer executes the method of claim 1 .

15. An artificial intelligence self-learning-based resting electrocardiogram analysis method, comprising:

receiving, at an analysis system, resting electrocardiogram data from a resting electrocardiogram monitoring device, and converting a data format of the resting ECG data into a preset standard data format by resampling, and performing a first filtering processing on converted resting ECG data in the preset standard data format;

performing, at the analysis system, heart beat detection processing on resting ECG data processed by the first filtering processing to identify multiple pieces of heart beat data comprised in the resting ECG data, each of which is corresponding to a heart beat cycle, comprising amplitude data and starting-ending time data of corresponding P wave, QRS complex and T wave; and lead parameters of the heart beat data are determined;

combining the heart beat data to generate heart beat time sequence data in the analysis system based on the lead parameters and time rules; and

generating heart beat analysis data in the analysis system according to the heart beat time sequence data;

performing, at the analysis system, feature extraction and analysis of an amplitude and time characterization data on the heart beat analysis data according to a trained heart beat classification model, to obtain primary classification information of the heart beat analysis data;

inputting the heart beat analysis data of particular heart beats in results of the primary classification information into a trained ST segment and T wave change model for identification, and determining ST segment and T wave evaluation information;

performing, at the analysis system, secondary classification processing on the heart beat analysis data according to resting ECG basic rule reference data, detailed feature information of the P wave and the T wave and the ST segment and T wave evaluation information under the primary classification information to finally obtain heart beat classification information;

according to the heart beat time sequence data and the heart beat classification information, performing, at the analysis system, heart beat waveform feature detection on the heart beat analysis data to determine feature heart beats;

determining, at the analysis system, detailed feature information of the QRS complex in the feature heart beats; determining the detailed feature information of the P wave and the T wave in the feature heart beats; and

calculating, at the analysis system, average waveform data and measurement values of the feature heart beats; and

for data of the heart beat classification information and the average waveform data of the feature heart beats, generating and outputting ECG event data, data of the measurement values and the average waveform data of the feature heart beats according to the resting ECG basic rule reference data and a classification standard of ECG Minnesota coding.